2024 Weeknote 39 : More UX, less AI

September 23 – 29

My “weeknotes” capture events, thoughts, and other items from the past week, mostly focused on work. Learn more about the weeknotes concept here.

I’m experimenting with a new format this week. I want to constrain my posts a bit and develop a repeatable theme. The new concept has 5 components:

• One Thought — A deeper exploration of something on my mind or something that happened.
• Five Notes — Quick hits of things going on or things I’d like to share.
• One Video — Something either entertaining or educational I’ve run across.
• Five Laughs — An abbreviated collection of Internet Funnies, usually picked up from Bluesky or Reddit.
• One Photo — A single photo, with caption, either from this week or sometime in the distant past.

So let’s try this out…

One Thought

It’s a good week when Recoding America’s Jennifer Pahlka replies to a comment you made on LinkedIn with a “yes, exactly.”

Pahlka shared an extended essay she shared here: AI Meets the Cascade of Rigidity which I read and then shared my reaction:

I hope the focus on AI in this piece is really just a “hook” to get readers to pay attention to the much bigger issues you clearly lay out. As you noted (after the AI exploration), government has massive barriers to adoption of all kinds of technologies, which in turn is driven by a lack of enabling behaviors, beliefs, and processes necessary to make all technology applications meaningful and useful. Most of us in government don’t need AI tools — we need value-focused product owners, human-centered service designers, and mission-driven technologists.

And for the next 10 years, none of them need to be deep AI experts. At most they could adopt AI tech infused into third-party software packages.

Buy all the AI tools you want, but if you don’t have all the other attitudes and behaviors fixed, it won’t matter. Indeed, AI could super-charge bureaucratic behaviors and barriers.

So… I’m going to keep AI on the back burner. We have some UX and CX ideas to get out into the minds of our government colleagues first.

In the original piece she focused a lot of energy on Artificial Intelligence (AI) and how it might interact with government at all levels, for solving problems, producing services, or accelerating processes. This is a topic lots of people are talking about, of course, and I’ve registered my distaste with the overly excitable coverage of AI in articles, webinars, and of course conferences. I see some uses, but not nearly as many in government as everyone would have you believe, precisely because government hasn’t done the basics yet — we don’t have well-groomed data sets or streamlined processes or even rational policies underlying our practices.

So I suspected the inclusion and the focus on AI was a bit of a “hook” to capture attention in the industry, because she flips to the themes of Recoding America in the last third. You can see the left turn here:

Fine-tuning between strict and loose controls is like obsessing over the safety features of cars while entirely neglecting driver education and licensing. The guardrails are in place, but the drivers don’t know how they work—nor how to actually drive. Responsible, effective use of AI will be a function of government’s competencies and capacities far more than its rules.

At which point she gets at some of the roots of the real problems in government digital services, and it’s not an AI problem or AI opportunity (boldface added):

People who understand both the systems in question and the possibilities of technology are far outnumbered by lawyers, compliance officers, and oversight bodies whose default is to stop rather than to go. They must spend far more time reporting (often to an absurd level of detail) on what they will do, what they are doing, and what they have just done, and seeking approvals from sometimes dozens of stakeholders, than actually building or deploying technology.

That atmosphere depresses the market for technologists and frustrates those of us in the field. I have to admit this is less of a problem in local government than in federal, but there are still elements of it (particularly in procurement).

So let’s keep our eyes on the prize: focusing on human-centered design, UX research, direct-to-public digital services, overall CX, and only deploy AI when it serves those interests first.

Five Notes

• Coaching Leaders: How to Respond When You Don’t Have Resources [podcast]
  Great advice that I learned the hard way over the years (and still sometimes forget) on how to say “no” without saying “no” or, oftentimes, how to say “yes” but set the terms of the “yes,” which are often far more negotiable than you think.
• the paradox of certainty [article]
  Not sure when this piece was originally posted, but it came to my attention this week and it’s a blockbuster for those of us trying to make a difference in a leadership role. It’s also deeply relevant to human-centered design (HCD), UX research, and other practices we need to employ when building solutions for real users in real life. Naturally, it comes from Michael Brennan at Civilla, an organization we’ve come to love in the last year, after following them and then taking a tour in July.
• The work is never just “the work” [article]
  This is a deep-dive highly-analytical piece that explores why it takes extra time to do all the work in-between the “actual” work. We so often short-change ourselves and each other when we ignore what Hack Your Bureaucracy calls “do the work outside the meeting.” Special thanks to Marcy Katz Jacobs for sharing this LinkedIn post by Stephanie Leary.
• We completed a “Digital Services Academy” series with Granicus this past Friday. It was a solid, if introductory, course delivered over 9 hours. Our entire GX Foundry team participated, and I think for several it was eye-opening. We have so much to learn, but also so much to share with our partners in the months and years to come. It really feels like we’re just getting started, but we’re getting started with the right things.
• Finally, we completed a homegrown “tabletop exercise” this past week between our customer relations folks and our project management folks, exploring how we work with customers, how we capture requirements, and how we hand off work between the teams. It was a good experience, but I have a lot of ideas for the next round. I may also need to find some kind of resource that teaches how to build these kinds of events.

One Video

This one is long, but amusing if you know Star Trek movies and specifically the poorly-reviewed Star Trek V, directed by William Shatner. It’s entertaining to see a well-executed takedown, but the host actually shares some ideas about how and why the movie isn’t that bad. Sure, it’s bad, but… it has some redeeming qualities.

Five Laughs

---

---

---

---

Okay this last one deserves a special award. This is a comedy Möbius strip.

One Photo